Implement Python Array API `asarray` function. by ysiraichi · Pull Request #60627 · pytorch/pytorch

ysiraichi · 2021-06-24T05:19:32Z

Stack from ghstack:

-> Implement Python Array API asarray function. #60627

In this PR, the core of frombuffer and fromDLPack onto tensor_new.cpp. asarray
uses such refactored functions for interpreting the object as a tensor. We follow the
Python Array API standard found:

https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray

Differential Revision: D31640510

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

facebook-github-bot · 2021-06-24T05:19:38Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/60627
📄 Preview docs built from this PR
📄 Preview C++ docs built from this PR
🔧 Opt-in to CIFlow to control what jobs run on your PRs

💊 CI failures summary and remediations

As of commit e1e6115 (more details on the Dr. CI page):

💚 💚 Looks good so far! There are no failures yet. 💚 💚

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 60b13a7 Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 681ab0d Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 8b8b22c Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 5ae565c Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 3bd82ef Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 78bce4f Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 20b3ced Pull Request resolved: #60627

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 71f40cb Pull Request resolved: #60627

ysiraichi · 2021-10-05T11:17:01Z

Hi, @anjali411 @mruberry. This is a friendly reminder about this PR. Do you have some time to take a look at it?

anjali411 · 2021-10-05T17:17:24Z

+    >>> a = torch.tensor([1, 2, 3])
+    >>> # Shares memory with tensor 'a'
+    >>> b = torch.asarray(a)
+    >>> a.data_ptr() == b.data_ptr()


okay yes so based on this, we should disable asarray for conjugated and negated tensors (see the comment above pointing to a PR that made a similar change for .numpy())

anjali411 · 2021-10-05T17:35:19Z

+def get_dtype_size(dtype):
+    return int(torch.empty((), dtype=dtype).element_size())
+
+class TestBufferProtocol(TestCase):


was this just verbatim moved from the file deleted above?

Yes. Reference: #60627 (comment)

anjali411

LGTM overall! It might be nice to add an OpInfo for asarray

cc. @mruberry for a final review

ysiraichi · 2021-10-07T10:46:00Z

Hi @anjali411. Thank you for your review.
I have tried to address your comments. Let me know if there's anything unclear.

It might be nice to add an OpInfo for asarray

I will do it in another PR. 😄

ysiraichi · 2021-10-11T09:47:58Z

Hi, @mruberry. This is a friendly reminder about this PR. Do you have some time to take a look at it?

mruberry · 2021-10-14T04:06:54Z

+    bool wrong_device = device.has_value() && device.value() != tensor.device();
+    bool wrong_dtype =
+        dtype.has_value() && dtype.value() != tensor.scalar_type();
+    bool needs_copying = !copy.has_value() && (wrong_device || wrong_dtype);


This was a tricky conditional to read because needs_copying would seem to be determined by whether the device or dtype is incorrect but the additional !copy.has_value() is confusing. Not worth revising the PR for

Ah. You are totally right.

mruberry

Hey @ysiraichi!

Thank you for your patience and pings. I finally had a chance to read through this PR again thoroughly (I tried several times but something always came up!). It's very good, and I appreciate @anjali411's concern but don't think this PR suffers from the issue. There are a few small tweaks I would like to suggest about readability and docs, but they're minor and I think we should land this as-is. Nice work, this is a challenging, detailed PR that required a great deal of effort, thoughtfulness, and attention to detail.

mruberry · 2021-10-14T04:22:29Z

@mruberry has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

ysiraichi · 2021-10-14T08:40:00Z

@mruberry @anjali411 Thank you for the review! I imagine it was a little bit tricky to review it, since I moved some code around in some non-straightforward way. Anyway, thank you for reviewing it.

mruberry · 2021-10-15T20:48:58Z

Ack! Looks like this just needs a rebase, @ysiraichi. Sorry about that. Just ping me once it's rebased.

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray Differential Revision: [D31640510](https://our.internmc.facebook.com/intern/diff/D31640510) [ghstack-poisoned]

pytorch-probot · 2021-10-16T08:01:43Z

CI Flow Status

⚛️ CI Flow

Ruleset - Version: v1
Ruleset - File: https://github.com/pytorch/pytorch/blob/aee9a2ba29edfee74506bc22e7f79d12a88cd258/.github/generated-ciflow-ruleset.json
PR ciflow labels: ciflow/default

Workflows	Labels (bold enabled)	Status
Triggered Workflows
linux-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/noarch`, `ciflow/xla`	✅ triggered
linux-vulkan-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/vulkan`	✅ triggered
linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-clang7-asan	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/sanitizers`	✅ triggered
linux-xenial-py3.6-clang7-onnx	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/onnx`	✅ triggered
linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-gcc7-bazel-test	`ciflow/all`, `ciflow/bazel`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
win-vs2019-cpu-py3	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/win`	✅ triggered
win-vs2019-cuda11.3-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/win`	✅ triggered
Skipped Workflows
libtorch-linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
libtorch-linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
linux-bionic-cuda10.2-py3.9-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
parallelnative-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped
periodic-libtorch-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`, `ciflow/slow`, `ciflow/slow-gradcheck`	🚫 skipped
periodic-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-win-vs2019-cuda11.1-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/scheduled`, `ciflow/win`	🚫 skipped
puretorch-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped

You can add a comment to the PR and tag @pytorchbot with the following commands:

# ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun

# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slow

For more information, please take a look at the CI Flow Wiki.

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 4816662 Pull Request resolved: #60627

ysiraichi · 2021-10-16T08:02:40Z

@mruberry Rebased!

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray Differential Revision: [D31640510](https://our.internmc.facebook.com/intern/diff/D31640510) [ghstack-poisoned]

In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray ghstack-source-id: 796f54f Pull Request resolved: #60627

pytorch-probot · 2021-10-16T10:05:34Z

CI Flow Status

⚛️ CI Flow

Ruleset - Version: v1
Ruleset - File: https://github.com/pytorch/pytorch/blob/e1e6115b2e184c1d04aeae31eb5031de9b4960d5/.github/generated-ciflow-ruleset.json
PR ciflow labels: ciflow/default

Workflows	Labels (bold enabled)	Status
Triggered Workflows
linux-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/noarch`, `ciflow/xla`	✅ triggered
linux-vulkan-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/vulkan`	✅ triggered
linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-clang7-asan	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/sanitizers`	✅ triggered
linux-xenial-py3.6-clang7-onnx	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/onnx`	✅ triggered
linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-gcc7-bazel-test	`ciflow/all`, `ciflow/bazel`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
win-vs2019-cpu-py3	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/win`	✅ triggered
win-vs2019-cuda11.3-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/win`	✅ triggered
Skipped Workflows
libtorch-linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
libtorch-linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
linux-bionic-cuda10.2-py3.9-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
parallelnative-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped
periodic-libtorch-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`, `ciflow/slow`, `ciflow/slow-gradcheck`	🚫 skipped
periodic-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-win-vs2019-cuda11.1-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/scheduled`, `ciflow/win`	🚫 skipped
puretorch-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped

You can add a comment to the PR and tag @pytorchbot with the following commands:

# ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun

# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slow

For more information, please take a look at the CI Flow Wiki.

pytorch-probot · 2021-10-16T10:05:34Z

CI Flow Status

⚛️ CI Flow

Ruleset - Version: v1
Ruleset - File: https://github.com/pytorch/pytorch/blob/e1e6115b2e184c1d04aeae31eb5031de9b4960d5/.github/generated-ciflow-ruleset.json
PR ciflow labels: ciflow/default

Workflows	Labels (bold enabled)	Status
Triggered Workflows
linux-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/noarch`, `ciflow/xla`	✅ triggered
linux-vulkan-bionic-py3.6-clang9	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/vulkan`	✅ triggered
linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-clang7-asan	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/sanitizers`	✅ triggered
linux-xenial-py3.6-clang7-onnx	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`, `ciflow/onnx`	✅ triggered
linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
linux-xenial-py3.6-gcc7-bazel-test	`ciflow/all`, `ciflow/bazel`, `ciflow/cpu`, `ciflow/default`, `ciflow/linux`	✅ triggered
win-vs2019-cpu-py3	`ciflow/all`, `ciflow/cpu`, `ciflow/default`, `ciflow/win`	✅ triggered
win-vs2019-cuda11.3-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/default`, `ciflow/win`	✅ triggered
Skipped Workflows
libtorch-linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
libtorch-linux-xenial-cuda11.3-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`	🚫 skipped
linux-bionic-cuda10.2-py3.9-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
linux-xenial-cuda10.2-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/slow`	🚫 skipped
parallelnative-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped
periodic-libtorch-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/libtorch`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`, `ciflow/slow`, `ciflow/slow-gradcheck`	🚫 skipped
periodic-linux-xenial-cuda11.1-py3.6-gcc7	`ciflow/all`, `ciflow/cuda`, `ciflow/linux`, `ciflow/scheduled`	🚫 skipped
periodic-win-vs2019-cuda11.1-py3	`ciflow/all`, `ciflow/cuda`, `ciflow/scheduled`, `ciflow/win`	🚫 skipped
puretorch-linux-xenial-py3.6-gcc5.4	`ciflow/all`, `ciflow/cpu`, `ciflow/linux`	🚫 skipped

You can add a comment to the PR and tag @pytorchbot with the following commands:

# ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun

# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slow

For more information, please take a look at the CI Flow Wiki.

mruberry · 2021-10-16T17:08:39Z

@mruberry has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

facebook-github-bot · 2021-10-17T04:13:03Z

@mruberry merged this pull request in 8854817.

Summary: Pull Request resolved: #60627 In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray Test Plan: Imported from OSS Reviewed By: H-Huang Differential Revision: D31640510 Pulled By: mruberry fbshipit-source-id: d0869e0d73cb50023d5866b001dac5d34ca30dfd

Summary: Pull Request resolved: pytorch#60627 In this PR, the core of `frombuffer` and `fromDLPack` onto _tensor_new.cpp_. `asarray` uses such refactored functions for interpreting the object as a tensor. We follow the Python Array API standard found: https://data-apis.org/array-api/latest/API_specification/creation_functions.html?highlight=asarray Test Plan: Imported from OSS Reviewed By: H-Huang Differential Revision: D31640510 Pulled By: mruberry fbshipit-source-id: d0869e0d73cb50023d5866b001dac5d34ca30dfd

ysiraichi mentioned this pull request Jun 24, 2021

Implement NumPy-like frombuffer tensor constructor. #59077

Closed

facebook-github-bot added the cla signed label Jun 24, 2021

pytorchbot added the open source label Jun 24, 2021

ysiraichi marked this pull request as ready for review July 8, 2021 13:49

ysiraichi requested review from albanD and soulitzer as code owners July 8, 2021 13:49

ysiraichi requested a review from mruberry July 8, 2021 13:50

anjali411 reviewed Oct 5, 2021

View reviewed changes

anjali411 approved these changes Oct 5, 2021

View reviewed changes

mruberry reviewed Oct 14, 2021

View reviewed changes

mruberry approved these changes Oct 14, 2021

View reviewed changes

pytorch-probot Bot added the ciflow/default label Oct 16, 2021

facebook-github-bot closed this in 8854817 Oct 17, 2021

facebook-github-bot added the Merged label Oct 17, 2021

facebook-github-bot deleted the gh/ysiraichi/13/head branch October 20, 2021 14:16

pmeier mentioned this pull request Jan 4, 2022

torch.asarray does not detect dtype of Python scalars #70591

Closed

Conversation

ysiraichi commented Jun 24, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Jun 24, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

Uh oh!

ysiraichi commented Oct 5, 2021

Uh oh!

anjali411 Oct 5, 2021

Choose a reason for hiding this comment

Uh oh!

anjali411 Oct 5, 2021

Choose a reason for hiding this comment

Uh oh!

ysiraichi Oct 7, 2021

Choose a reason for hiding this comment

Uh oh!

anjali411 left a comment

Choose a reason for hiding this comment

Uh oh!

ysiraichi commented Oct 7, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ysiraichi commented Oct 11, 2021

Uh oh!

mruberry Oct 14, 2021

Choose a reason for hiding this comment

Uh oh!

ysiraichi Oct 14, 2021

Choose a reason for hiding this comment

Uh oh!

mruberry left a comment

Choose a reason for hiding this comment

Uh oh!

mruberry commented Oct 14, 2021

Uh oh!

ysiraichi commented Oct 14, 2021

Uh oh!

mruberry commented Oct 15, 2021

Uh oh!

pytorch-probot Bot commented Oct 16, 2021

⚛️ CI Flow

Uh oh!

ysiraichi commented Oct 16, 2021

Uh oh!

pytorch-probot Bot commented Oct 16, 2021

⚛️ CI Flow

Uh oh!

pytorch-probot Bot commented Oct 16, 2021

⚛️ CI Flow

Uh oh!

mruberry commented Oct 16, 2021

Uh oh!

facebook-github-bot commented Oct 17, 2021

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

ysiraichi commented Jun 24, 2021 •

edited

Loading

facebook-github-bot commented Jun 24, 2021 •

edited

Loading

ysiraichi commented Oct 7, 2021 •

edited

Loading